Noustiny
An agent native video creation pipeline that runs on top of Hermes Agent
What is it?
What it is
Noustiny is an interactive story‑building platform where a user drops a seed prompt and the system automatically expands a branching narrative tree. Each beat is authored by a Hermes Agent that proposes options, writes inserted beats, audits continuity, and patches downstream content. The platform also generates IP‑free character portraits, voice‑cloned narration from YouTube or text, and renders the chosen branch into a video ready for posting.
Why it exists
It exists to make creative storytelling effortless and collaborative: users can explore countless "what if" branches without manual writing, while the agent handles prose, visual consistency, voice synthesis, and rendering, turning imagination into shareable media.
Who should use it
creative writersstorytellersgame designerscontent creatorstech-savvy users interested in AI tools
Who should avoid it
users seeking simple tools without technical setupthose uninterested in creative storytelling or multimedia elements
How it works
A quick walkthrough in plain English
How Noustiny works
Step 1 of 3
You interact with it
Open Noustiny, send a request, or connect it to your stack.
Features
Advantages
- Innovative autonomous storytelling with Hermes Agent
- Seamless integration of voice, visuals, and narrative
- Automatic story coherence after splices
- User-friendly gesture-based interface
- Open-source and customizable for creative projects
- Supports multiple media formats (video, audiobook)
- Strong character consistency across the story
- Leverages advanced NLP and AI tools
Disadvantages
- Requires technical setup (WSL, Windows mirror)
- Complex system with multiple dependencies
- No explicit license mentioned
- Potential learning curve for new users
- Reliance on specific external services (ElevenLabs, YouTube)
- Limited documentation for some tools
- May require advanced NLP/creative skills
Installation
FAQ
How do I start a new story in Noustiny?
Open the canvas and type a seed prompt. The system will generate a canon‑tracked, live branching story canvas based on that seed.
What’s the difference between the square + and round + buttons on the canvas?
The square + (gold parallelogram) splices a new beat into the canon; it triggers the narrative‑writer‑assist and a cascade that re‑harmonises downstream beats. The round + (dashed circle) adds a what‑if sibling branch that doesn’t touch the canon and only produces fresh next options.
How does the render process work?
Press Render and the agent chains four Hermes tools in one chat completion: narration‑voice‑director → voice_sample_builder → voice_clone_synthesize → noustiny_storybook. The result is a video (landscape audiobook or vertical reel) ready to post.
How are character portraits kept consistent across the story?
At seed time a character sheet is built with up to four IP‑free portraits. Every beat that mentions a character resolves aliases (e.g., “Mr. Stark” → “Tony Stark”) and pulls the same portrait reference, ensuring a single face per character throughout the tree and the rendered video.
How do I add a YouTube URL for voice cloning?
Drop a YouTube URL (or 11‑character ID or free text) into the canvas. The voice_sample_builder tool downloads the audio via yt‑dlp, normalises it to 24 kHz mono PCM, and provides a reference WAV for the voice_clone_synthesize step.
Featured in Videos
YouTube tutorials and walkthroughs for Noustiny
Alternatives
Similar projects ranked by category, topics, and text overlap.
